Akademik Veri Yönetim Sistemi
TÜBİTAK Projesi, 2014 - 2016
"This project is on evacuation planning, more precisely, on selecting shelter sites and assigning the traffic to routes optimally to evacuate a disaster region as quickly as possible. In this project we develop a model that assigns evacuees to the nearest shelter sites by assigning them to the shortest paths to those shelter sites, shortest and nearest with a given degree of tolerance, and by optimally locating the shelter sites so that the total evacuation time is minimized. Our model differs from location-allocation models in the evacuation literature in that we decide the optimal locations of shelter sites in a SO manner while assigning the evacuees to shelters and to routes from the point of view of the evacuees. As the travel time on a road segment is often modeled as a nonlinear function of the flow on the segment, the resulting model is be a nonlinear mixed integer programming model. In the first part of the project, we develop solution methods that can handle practical size problems by using second order cone programming techniques. In the second part, we give a comparison of the UE, SO and shortest path approaches in terms of several performace measures such as total evacuation time, price of anarchy, fairness, efficiency, and maximum latency. In the third part of the project, we extend these results to an environment with uncertain parameters and consider a two stage problem where shelter location decisions are made in the first stage and the routing decisions are made in the second stage. Finally, we extend these results into a setting where the loads assigned to shelters are balanced. Since the problems we are taking in hand in this project are new, we believe the work done will fill a gap in the literature. We plan to submit the results of this study to international journals and share them in conferences. It is our long term aim that the graduate students who will work on this project be qualified researchers in the area of evacuation, traffic management and facility location problems, and take part in studies of evacuation planning and management in our country and abroad and that the results produced be used in practice. We have no doubt that the studies will start new projects since they bring all together different interesting topics such as traffic assignment, facility location, second order conic programming, two stage stochastic programming and load balancing."